iocell list recommends a migration to ucl, but fails to execute it. #52
Comments
|
Hello, Can you please update to a recent version (2.2.0 has been in ports/pkg since february) and run iocell as root? IIRC the switch to UCL was a long time ago - what OS version are you running and/or from which version did you upgrade prior to getting that error? Re-running 'activate' on an already activated pool shouldn't cause any issues, IIRC iocell just skips any dataset that's already there, but not 100% sure now. I have a test system at hand where I could try that tomorrow. Regards, |
This what puzzled me too. I scanned the commit history and saw commits related to UCL from several years ago. Anyway, I'm on FreeBSD 13.3, and I think things got messed up on my system. however |
|
Actually, 2.1.2 is hardcoded, from what I can tell. https://github.com/bartekrutkowski/iocell/blob/master/iocell#L34 |
Did you just recently upgrade to 13.3 (possibly from a much older release + iocell version)? Also, can you try running iocell as root instead of via sudo - sudo has various implications to what it does (or doesn't do) to the environment variables and shell; e.g. it might be possible that shellouts don't inherit the elevated privileges and hence iocell fails to access some paths/files. What makes this worse is that sudo has become a fast moving (and bloated) target and often changes behavior. I really can't reproduce this error. Also what puzzles me is that iocell is looking for '/iocell/jails/config' - that config file however should be within each jails dataset/path... So this looks like iocell gets an empty jail uuid from somewhere and hence all the paths are breaking... |
True, that's actually bad, at least because it wasn't updated. I'm not sure if a PR against the development branch makes sense, as the release/tag couldn't be updated that way - at least I assume that from my (very limited) understanding how git releases/tags work... @bartekrutkowski how can/should this be updated for the 2.2.0 tag? Or could some of the PRs be merged and we move on to the next tag and update the version string at the same time? |
No recent world updates or upgrades. But this all started after I upgraded ports. This is when I think iocell upgraded from actual 2.1.2 to 2.2.0. Looks like the port got updated around Feb-Mar time range. |
Ok, did that, and after that when I run but this error is still there. However, when I run `iocell list' under root now, i'm still seeing all the errors from the 1st post. |
|
|
OK, nothing suspicios with the datasets, however where does the mount |
|
Good catch, I didn't notice it before. It doesn't seem to have anything in it. I think me trying to fix it on my own lead to me somehow creating it. I removed it with |
Hi,
I've recently hit this strange behavior of iocell that I haven't been able to resolve.
Version: iocell 2.1.2 (2017-06-17)
When I run
listit reports this errorFollowing the recommendation is failing with the following output:
Additional information:
Prior to getting these errors I had iocell getting removed during one of the port updates. I didn't catch what lead to iocell removal though. iocell was reinstalled, and I ran
iocell activateto point it to an already existing pool. After doing that I see all the jails I have, but now with this ucl errors.Make sure to follow and check these boxes before submitting an issue! Thank you.
iocell versioniocelldevelopif using the stable release ormasterbranch. Skipped, due to stability concerns of the develop branch, called out on the main page.Thank you.
The text was updated successfully, but these errors were encountered: